The goal of the proposed research is to conduct the strictest test to date of the hypothesis that fear can be reduced without awareness. Two experiments will test this hypothesis with respect to brain and behavior in order to investigate novel processes of non-conscious fear reduction and their potential practical utility. This hypothesis challenges the prevailing clinical belief that a feared object must be directly confronted with full awareness in order to reduce fear of it. The experimental paradigm, very brief exposure, consists of a continuous series of backward masked phobic images (each for duration of 33-ms). A series of experiments has shown that VBE reduces avoidance of a feared object (Siegel, Anderson & Han, 2011; Weinberger, et al., 2011; Siegel & Weinberger, 2009), more than clearly visible exposure (CVE) to the same stimuli does (Siegel & Weinberger, 2012), and that the avoidance- reducing effect lasts for at least a year (Siegel & Warren, in press). An fMRI and behavioral experiment are designed to identify the learning mechanisms of VBE and link them to its effect on phobic avoidance. The fMRI experiment will be conducted at the MRI Laboratory of the New York State Psychiatric Institute. Twenty-six spider-phobic and twenty-six non-phobic participants will be recruited from liberal arts and science courses at Purchase College/SUNY with a widely used questionnaire, a Behavioral Avoidance Test (BAT), and the Structured Clinical Interview for DSM-IV Diagnosis. Each of three types of exposure will be administered within-subjects: very brief (33.4-ms SOA) and clearly visible (119-ms SOA) images of spiders, and very brief images of flowers (control). Subjective fear is rated periodically throughout. VBE is expected to habituate amygdala activation in the absence of the experience of fear, whereas CVE will not habituate fear responses and induce subjective fear. The behavioral experiment will be conducted in the Principal Investigator's lab at Purchase College/SUNY. 80 spider-phobic participants - 26 from the fMRI experiment, and 74 others - will be recruited in the same manner. Based on baseline levels of avoidance of a live tarantula, they will be randomly assigned to exposure to very brief (33-ms SOA) images of either spiders or flowers (control group). These exposures will be administered each of three consecutive weeks, followed by in vivo exposure to a live tarantula. Subjective fear will be measured just before and after these exposures. Reduction of amygdala activation by VBE in the fMRI experiment will predict reduction of avoidance of the tarantula. VBE will reduce avoidance of a live tarantula after the first session. Control participants will not make the same gain until the 3rd session of n vivo exposure. The two subsequent VBE sessions will result in further gains. The proposed research will lay the groundwork for an R01 Proposal to further investigate neurobiological mechanisms of very brief exposure and test its effects on clinical populations.